Handle unprintable characters more gracefully

* Skip UTF-8 byte order mark at beginning of file
* Error on other unexpected unprintable characters
This commit is contained in:
dbrotz
2019-06-29 12:50:41 -07:00
parent 1decf5d0d4
commit 484d15dbb2
3 changed files with 32 additions and 12 deletions

View File

@@ -38,6 +38,8 @@ struct sLexString {
#define SAFETYMARGIN 1024 #define SAFETYMARGIN 1024
#define BOM_SIZE 3
struct sLexFloat tLexFloat[32]; struct sLexFloat tLexFloat[32];
struct sLexString *tLexHash[LEXHASHSIZE]; struct sLexString *tLexHash[LEXHASHSIZE];
YY_BUFFER_STATE pCurrentBuffer; YY_BUFFER_STATE pCurrentBuffer;
@@ -49,6 +51,9 @@ uint32_t tFloatingChars[256];
uint32_t nFloating; uint32_t nFloating;
enum eLexerState lexerstate = LEX_STATE_NORMAL; enum eLexerState lexerstate = LEX_STATE_NORMAL;
/* UTF-8 byte order mark */
static const unsigned char bom[BOM_SIZE] = { 0xEF, 0xBB, 0xBF };
void upperstring(char *s) void upperstring(char *s)
{ {
while (*s) { while (*s) {
@@ -120,11 +125,11 @@ void yy_delete_buffer(YY_BUFFER_STATE buf)
*/ */
static void yy_buffer_append(YY_BUFFER_STATE buf, uint32_t capacity, char c) static void yy_buffer_append(YY_BUFFER_STATE buf, uint32_t capacity, char c)
{ {
assert(buf->pBuffer[buf->nBufferSize] == 0); assert(buf->pBufferStart[buf->nBufferSize] == 0);
assert(buf->nBufferSize + 1 < capacity); assert(buf->nBufferSize + 1 < capacity);
buf->pBuffer[buf->nBufferSize++] = c; buf->pBufferStart[buf->nBufferSize++] = c;
buf->pBuffer[buf->nBufferSize] = 0; buf->pBufferStart[buf->nBufferSize] = 0;
} }
YY_BUFFER_STATE yy_scan_bytes(char *mem, uint32_t size) YY_BUFFER_STATE yy_scan_bytes(char *mem, uint32_t size)
@@ -221,6 +226,11 @@ YY_BUFFER_STATE yy_create_buffer(FILE *f)
*/ */
capacity += 3; capacity += 3;
/* Skip UTF-8 byte order mark. */
if (pBuffer->nBufferSize >= BOM_SIZE
&& !memcmp(pBuffer->pBuffer, bom, BOM_SIZE))
pBuffer->pBuffer += BOM_SIZE;
/* Convert all line endings to LF and spaces */ /* Convert all line endings to LF and spaces */
char *mem = pBuffer->pBuffer; char *mem = pBuffer->pBuffer;
@@ -281,7 +291,7 @@ YY_BUFFER_STATE yy_create_buffer(FILE *f)
} }
/* Add newline if file doesn't end with one */ /* Add newline if file doesn't end with one */
if (size == 0 || pBuffer->pBuffer[size - 1] != '\n') if (size == 0 || pBuffer->pBufferStart[size - 1] != '\n')
yy_buffer_append(pBuffer, capacity, '\n'); yy_buffer_append(pBuffer, capacity, '\n');
/* Add newline if \ will eat the last newline */ /* Add newline if \ will eat the last newline */
@@ -289,10 +299,10 @@ YY_BUFFER_STATE yy_create_buffer(FILE *f)
size_t pos = pBuffer->nBufferSize - 2; size_t pos = pBuffer->nBufferSize - 2;
/* Skip spaces */ /* Skip spaces */
while (pos > 0 && pBuffer->pBuffer[pos] == ' ') while (pos > 0 && pBuffer->pBufferStart[pos] == ' ')
pos--; pos--;
if (pBuffer->pBuffer[pos] == '\\') if (pBuffer->pBufferStart[pos] == '\\')
yy_buffer_append(pBuffer, capacity, '\n'); yy_buffer_append(pBuffer, capacity, '\n');
} }
@@ -783,10 +793,22 @@ scanagain:
* numeric literal, string, or bracketed symbol, so just return * numeric literal, string, or bracketed symbol, so just return
* the ASCII character. * the ASCII character.
*/ */
if (*pLexBuffer == '\n') unsigned char ch = *pLexBuffer++;
if (ch == '\n')
AtLineStart = 1; AtLineStart = 1;
return *pLexBuffer++; /*
* Check for invalid unprintable characters.
* They may not be readily apparent in a text editor,
* so this is useful for identifying encoding problems.
*/
if (ch != 0
&& ch != '\n'
&& !(ch >= 0x20 && ch <= 0x7E))
fatalerror("Found garbage character: 0x%02X", ch);
return ch;
} }
if (pLongestFixed == NULL || nFloatLen > pLongestFixed->nNameLength) { if (pLongestFixed == NULL || nFloatLen > pLongestFixed->nNameLength) {

View File

@@ -1,3 +1,2 @@
ERROR: garbage_char.asm(1): ERROR: garbage_char.asm(1):
syntax error Found garbage character: 0xFF
error: Assembly aborted (1 errors)!

View File

@@ -1,3 +1,2 @@
ERROR: -(1): ERROR: -(1):
syntax error Found garbage character: 0xFF
error: Assembly aborted (1 errors)!